Preparing for 6.3 Release

Ken Moffat ken at linuxfromscratch.org
Fri Jun 8 17:55:46 PDT 2007


On Fri, Jun 08, 2007 at 02:23:53PM -0700, Dan Nicholson wrote:
> So, I think it's about time we start pushing out a 6.3 release. What
> do you guys think? Here are the outstanding issues I can think of.
> 
> * Linux-2.6.21: Yesterday I read this blog post from the Fedora kernel
> maintainer,
>   Dave Jones:
> 
>    http://kernelslacker.livejournal.com/79957.html
> 
>    Doesn't exactly make 2.6.21 sound like the appealing kernel to base
> our stable
>    release on. I'm not the expert here, but it may be better to wait
> for 2.6.22. Food
>    for thought, anyway.
 And 2.6.22 will be better ?  From my reading, much of the problem
for fedora is because they moved to libata - one of my boxes is
using that, and has been for about 6 months, I think you are too.
So, for some of us it works, and anyway there is no requirement to
move from the old ide drivers.  Actually, I think recent kernels are
a lot better, I just hope 2.6.22 gets rid of that offensive message
about updating 'shutdown' (maybe it has already gone, I haven't
tested since -rc2).  I know Linus is unpredictable, but I think .22
is probably at least two weeks away, plus a couple more weeks for
stable updates to fix whatever major bugs show up.
> 
>  * Docbook XML/XSL changes: Manuel, how's that coming along? I haven't really
>    been paying attention to this lately.
> 
>  * Udev rules: Alexander has posted in a couple other places that we have broken
>    rules for DVB and floppy device setup. Alexander/Bryan, could you
> guys look at
>    our ruleset and make sure they do everything we want them to?
> 
>  * Bootscripts: There are a couple changes that I'd like to get in.
> Nothing critical.
>    I'll have a separate post on that.
> 
> Anything else?
> 
 If we don't release soon, everybody will be expecting updates to
gcc and glibc.  For x86 I have no opinion on gcc-4.2.0 (it's beaten
me to a pulp on ppc64) and at the moment I don't know if the damage
from glibc acquiring its own futimens() (possible spelling error)
extends into blfs.  I'm just about to try a build of my own with
newer non-toolchain packages, but I'd be happy to stick with what we
have plus 2.6.21.5 if that means we can get something out soon - I'm
starting to get offended by people mentioning gcc-4.0 in support
problems ;-)

ĸen
-- 
das eine Mal als Tragödie, das andere Mal als Farce



More information about the lfs-dev mailing list